92 results about "Error tolerance" patented technology

The present disclosure relates to an extended reality (XR)-based supervision assisting device, which finds a construction error by comparing a real design object photographed by a camera of the supervision assisting device with building information modelling (BIM) design data, and corrects the BIM design data to fit the real design object when the construction error between the real design object and the BIM design data exceeds a preset error tolerance range, thereby adjusting the BIM design data according to the construction error.

The invention relates to the technical field of vehicle braking control, in particular to a self-adaptive error tolerance automatic emergency braking system and method, and provides a self-adaptive error tolerance automatic emergency braking system and a forward collision early warning system. The method aims at solving the accuracy problem caused by sensor errors in an existing system. The method specifically comprises the following steps: introducing a first-order error propagation model to quantify a key safety index error; comprising the steps of collision time real-time estimation based on a forward sensor, error propagation and tolerance analysis and FCW / AEB trigger logic design introducing a fault-tolerant band. In the design, the maximum tolerance error is set, and the sensor index requirement is reversely deduced, so that the robustness and safety of the system are improved, a quantifiable fault-tolerant boundary is provided for the AEB system, the application effect of the system in a complex road environment is improved, and the method is particularly suitable for high-risk scenes such as expressways and urban expressways and has wide application prospects. And the fault-tolerant capability and the practical application effect of the system are enhanced.

The invention relates to the technical field of structural design and optimization, and particularly discloses a structural error tolerance model construction method for compartment connection positioning, which comprises the following steps of: obtaining real point cloud data of a bolt connection pair through three-dimensional laser scanning, and realizing parameterized digital geometric modeling in combination with a deep neural network; material attributes are given, and self-adaptive grid division is carried out; generating a dynamic load sequence based on a vehicle driving condition, applying the dynamic load sequence to the model, and extracting full-field displacement vector data by using a digital image correlation technology; further identifying a micro-slip area of the contact interface, and calculating an energy dissipation index of the micro-slip area; a quantitative mapping relation between energy dissipation and fatigue life is established through experimental calibration, key design parameters are iteratively optimized by adopting a pattern search method with the target life as a constraint, and finally a connection structure scheme meeting the durability requirement is output.

Provided is a method for memory pool management, the method including receiving, by a memory-pool manager, a memory request from an application, the memory-pool manager being communicatively coupled to a memory pool including a first memory module, of a first type, and a second memory module, of a second type, the first type being different from the second type, determining, by the memory-pool manager, based on the memory request, an error tolerance associated with the application, and allocating a memory space from the first memory module or from the second memory module to the application based on the error tolerance.

Disclosed in the embodiments of the present application are a data storage method and apparatus, a device, and a non-transitory readable storage medium. According to the embodiments of the present application, F parity blocks occupying blank storage areas are added to a RAID system, which not only increases error tolerance but also eliminates the need for additional parity disks. During data storage, corresponding original check codes are solved according to an original encoding method, and additional check codes located on the parity blocks are solved simultaneously. Thus, it is possible to recover more erroneous data based on the original check codes and the additional check codes. When an error occurs in a disk in which any block in a stripe is located and data loss occurs in other blocks of the same stripe, the corresponding data can still be recovered.

The invention discloses a static sounding probe rapid calibration and data deviation rectification method, and relates to the technical field of data deviation rectification, and the method comprises the following steps: carrying out a primary static sounding test, and recording initial calibration data of a probe; performing comparative analysis on initial calibration data obtained in the initial static sounding test and theoretical data of a set standard to obtain an error range and a deviation type of the probe; according to the error range and the deviation type of the probe, a sensor of the static sounding probe is adjusted and calibrated; in the actual test process, static sounding test data are collected in real time and compared with a preset error tolerance range, whether the current data exceed the error range or not is judged, and if the current data exceed the error range, a data deviation correction algorithm is started; and after data correction, further correcting and adjusting the probe through a feedback mechanism and verifying whether the data before and after correction meet a predetermined standard, and if not, continuing to adjust.

The invention relates to the technical field of printed circuit manufacturing, in particular to a drilling control method, system and equipment suitable for a circuit board, distance characteristics are obtained according to the position of a next drill hole of a current drill hole in the circuit board and the distance between each circuit in the circuit board, and the drilling error tolerance of the next drill hole is obtained according to the distance characteristics; according to the method, the initial drilling feeding speed is adjusted according to the drilling error tolerance and the drilling protrusion degree, the feeding speed matched with the next drilling hole can be obtained, drilling is conducted at the feeding speed, the drilling precision can be improved, the drilling quality can be ensured, and therefore influences on the performance of the circuit board are avoided; according to the drilling error tolerance and the drilling deviation degree, the maximum tool retracting speed of the drilling is adjusted, the target tool retracting speed of the next drilling is obtained, the drilling precision can be improved, secondary damage to the drilling during tool retracting is avoided, the drilling quality is ensured, and therefore the influence on the performance of a circuit board is avoided.

The invention discloses a large language model agent system and method and related equipment. The large language model agent system comprises an input analyzer module, a strategy generation module, an execution controller module, a task result evaluation module, an reflection module and a strategy rewriting module, and a self-adaptive optimization closed loop for task execution is formed through the reflection module and the strategy rewriting module. Wherein the reflection module can generate a report containing failure reasons, and the problem of failure diagnosis is solved; the strategy rewriting module optimizes a plan according to a report, adjusts tool calling, and realizes self-learning and strategy reconstruction; the closed-loop design forms a complete cognitive process, and an input analyzer guarantees the stability of prompt. According to the system, the task completion rate, the execution stability and the scene adaptability are remarkably improved, the robustness and the error tolerance in a multi-step complex task are particularly enhanced, and the key blank of an existing language model agent system in the aspects of self-reflection and self-optimization is filled.

A system, method, and computer-program product includes receiving a plurality of vector embeddings having an initial dimensionality and projecting the plurality of vector embeddings into lower-dimensional spaces using at least two different dimension reduction algorithms to generate corresponding sets of projected vector embeddings. Each set of projected embeddings may be quantized and nearest neighbors for the original embeddings and for each quantized set of projected embeddings may be calculated. Additionally, a neighbor preservation metric may be evaluated for each quantized set by comparing its nearest neighbors to those of the original embeddings. Based on the neighbor preservation metrics and a predefined error tolerance, an optimal compression configuration may be selected.

The present disclosure provides a method of forming N-type and P-type source / drain features using one patterned mask and one self-aligned mask to increase windows of error tolerance and provide flexibilities for source / drain features of various shapes and / or volumes. The present disclosure also includes forming a trench between neighboring source / drain features to remove bridging between the neighboring source / drain features. In some embodiments, the trenches between the source / drain features are formed by etching from the backside of the substrate.

Disclosed in the embodiments of the present application are a data storage method and apparatus, a device, and a non-transitory readable storage medium. According to the embodiments of the present application, F parity blocks occupying blank storage areas are added to a RAID system, which not only increases error tolerance but also eliminates the need for additional parity disks. During data storage, corresponding original check codes are solved according to an original encoding method, and additional check codes located on the parity blocks are solved simultaneously. Thus, it is possible to recover more erroneous data based on the original check codes and the additional check codes. When an error occurs in a disk in which any block in a stripe is located and data loss occurs in other blocks of the same stripe, the corresponding data can still be recovered.

Data security for a dataset in interactive query or operation from users regarding data stored in the dataset is provided. A method for providing data security for the dataset in secure data computation and communication includes generating a query result corresponding to the user query for the dataset, determining a magnitude range of the query result, and generating an amount of random noise data based on the magnitude range. The amount of random noise data is calibrated by adjusting at least one of a first tunable parameter within a first range and a second tunable parameter within a second range of a differential privacy (DP) configuration. A noise-laden query result is generated by applying the noise data to the query result to satisfy an error tolerance level.

The invention discloses a target data selection method based on an approximate Shapley value, electronic equipment and a medium. The target data selection method comprises the steps of obtaining a ratio of a source data set to a target utility, and calculating a total utility value and a target utility value. And screening to obtain an extended sample and a feature set. And sample optimization: carrying out Monte Carlo sampling on the expanded sample set to generate random arrangement, forming a plurality of third subsets and calculating utility values of the third subsets. And after a sample Shapley value is calculated, selecting a subset with the minimum error to carry out iterative sample replacement. And if the error reduction degree is smaller than a second threshold value, re-sampling, otherwise, taking a subset of which the error is smaller than a third threshold value as an initial candidate. And if the final error still exceeds the tolerance threshold, turning to feature optimization. And feature optimization: the process is similar to sample optimization, and feature iteration and replacement are guided through a Shapley value. And alternately optimizing the sample subsets and the feature subsets until the utility value error of the combined candidate set (the candidate sample subsets + the candidate feature subsets) is smaller than an error tolerance threshold.

The invention relates to the technical field of numerical control machining, in particular to a CNC closed-loop control method based on a dynamic feature model in a CAD / CAM environment. The method comprises the following steps: extracting a tolerance mark value of each processing surface in a CAD model, and mapping the tolerance mark value into an error tolerance budget of a corresponding track segment; the geometric curvature and the feeding speed set value of each track section in the machining program are analyzed from the CAM system, and the predicted contour error of each track section at the current speed is calculated according to the geometric curvature and the feeding speed set value; comparing the predicted contour error of each trajectory segment with the corresponding error tolerance budget, and calculating a tolerance margin; and marking the track section of which the tolerance margin is a negative value as an error overrun section. According to the method, tolerance transfer is carried out from the tolerance surplus section to the error over-limit section, so that error tolerance resources are redistributed among different track sections, and the stability and adaptability of processing beat control under a complex processing path are enhanced.

The present application discloses a communication method and apparatus. The method comprises: a network device sends a first data block to a terminal, the first data block corresponding to a first check code; and the terminal sends first information to the network device, the first information being used for indicating that the first data block has been successfully received or that the first data block has not been successfully received, the first information being determined on the basis of a first error tolerance rate and a check result corresponding to the first data block, and the check result being determined on the basis of the first check code. The described method can reduce excessive retransmissions caused by unsuccessful transmission of a first data block, thereby increasing system throughput of a communication system.

In a first correction process, a correction value T(0) is set as a median value between minimum and maximum correction values Tmin(0) and Tmax(0). A provisional adjustment value B=T(0)+Bini is input to the circuit section. A signal value output from the circuit section in accordance with the provisional adjustment value B is obtained by an A / D converter, and the magnitude relationship between the signal value and a target value is determined. When the signal value is smaller than the target value, a value that is smaller than the correction value T(0) by an error tolerance is set as a new minimum correction value Tmin(1). The median value between the minimum correction value Tmin(1) and the maximum correction value Tmax(1)=Tmax(0) is set as a next correction value T(1). In the next correction process, the provisional adjustment value B=T(1)+Bini is input to the circuit section.

The present invention discloses a method for guiding in the later stage of coaxial assembly based on machine vision, which is used to improve the degree of automation of assembly operations. The steps include: installing a camera at the front of the active end of the assembly to collect operation images, and pasting a target on the passive end of the assembly for end point identification, so as to form an assembly vision end navigation system; preprocessing the collected images to obtain a binary image containing the target image area; constructing a multi-level filter to filter the binary image to obtain a binary image with only target information; performing sub-pixel accuracy calculation on the target binary image to obtain the pixel position coordinates of the target; calibrating the camera to obtain the relationship between pixels and physical movement distances; calculating the offset distance for assembly navigation; after the navigation is completed, collecting the scene images again for error analysis. If the requirements are met, the assembly navigation is ended. If not, compensation is performed through step-by-step movement until the error tolerance requirements are met.

The utility model relates to a glass glue coating device with high error tolerance, including a connecting flange connected to the working end of a manipulator, the connecting flange is installed in cooperation with a first buffer unit through a connecting seat, the bottom of the first buffer unit is installed in cooperation with a glue gun through a second buffer unit, a pipe joint is installed in cooperation with the glue gun, and a felt block is installed in cooperation with the working end of the glue gun; adhesive is poured into the glue gun through the pipe joint, and the adhesive in the glue gun is applied to the edge of the glass through the felt block. By setting the first buffer unit and the second buffer unit, the action of the manipulator can be compensated, so that the felt block can fit the edge of the glass, the utility model has good practicability, and the production efficiency can be effectively improved.

The invention relates to a digital multi-bit RRAM calculation method and device for protecting high-weight data, and the method comprises the steps: carrying out the separated processing of the output of a multi-bit RRAM, storing a plurality of pieces of weight data in one RRAM, and carrying out the separation protection or direct output to protect the high order, so as to guarantee the accuracy of the data; according to the method, the fault-tolerant rate of neural network convolution is combined, the calculation power consumption is greatly reduced in the test process, the accuracy and the power consumption are balanced, the multiplication and addition calculation is more flexible, and the data reliability is greatly ensured. Therefore, the problems that due to the fact that the output of the RRAM is very dense and instability exists in the RRAM, drifting possibly exists, and part of data is prone to being flipped are solved.

The invention discloses an OPC verification method. The method comprises the following steps: step 1, carrying out region division on a layout and obtaining selected regions; and 2, dividing the pattern area of the selected area into a sub-regular pattern and a regular pattern based on a design rule. And step 3, marking the sub-regular graph to form a sub-regular graph mark. A halo area with the diameter being a first value is formed by expanding with the mark center of the sub-regular pattern as the original point and serves as a first mark area layer of the sub-regular pattern. And taking the non-taking area of the first marking area layer as a second marking area layer of the regular pattern. And 4, performing OPC verification on the patterns of the first and second mark area layers respectively. The item of OPC verification of the first mark area layer comprises first critical dimension inspection; the first critical dimension inspection adopts hierarchical inspection, and the first error tolerance of each layer of inspection is determined according to a set value of a design rule multiplied by a corresponding percentage. According to the invention, the inspection efficiency can be improved, omission is avoided, and the product yield can be improved.

The invention provides a distributed new energy asset chaining method and system based on error control, and belongs to the field of block chain technology and asset digitization. According to the method, operation data under a chain are obtained, reported through a plurality of independent oracle machine nodes and aggregated on the chain to form a consensus state, and then the consensus state is verified according to error tolerance parameters. If the verification is passed, updating a corresponding real world asset token state; and if the verification is not passed, the token is placed in a function-limited risk state, and the chain interaction of the token is limited. According to the method, the data error is converted into a manageable state on a chain instead of a system exception, so that the problem that the availability of the system is reduced due to the data error in the prior art is solved, and the robustness of the system is improved.

The present application provides a method for error-tolerant translation, a method for training an error-tolerant translation model, and an apparatus. The error-tolerant translation method comprises: identifying multiple candidate sentences corresponding to a source language sentence; determining a fused candidate sentence based on the multiple candidate sentences, the fused candidate sentence comprising multiple candidate words corresponding to the first word in the source language sentence and label information for indicating the multiple candidate words; encoding, pre-selecting, and decoding the fused candidate sentence using an error-tolerant translation model to obtain a target language sentence. The present application can improve the error tolerance and real-time performance of translation.

The present application relates to a method, device, equipment and storage medium for identifying the reading of a pointer instrument; in the present application, it is first necessary to respectively identify the pointer line segment, key scale points, text detection frame and text recognition results in the instrument image, and determine the scale value of each key scale point by matching, which can improve the recognition accuracy; and, when determining the final reading, the present application determines multiple initial readings by combining the scale values ​​of any two key scale points, and determines the final reading based on all combined initial readings; this method can reduce excessive reliance on the accuracy of specific scale point positions and scale values, thereby improving the fault tolerance rate, so that the present solution can still accurately identify the instrument reading in the presence of interference in the instrument image, unclear scale values, pointer blocking the scale value and other interference conditions, making the present solution more robust and adaptable when identifying the instrument reading.

An error-tolerant communication method and apparatus. The method comprises: an access network device sends a plurality of data units to a terminal, the plurality of data units corresponding to a first feature stream; and the terminal sends feedback information for the plurality of data units on the basis of the transmission condition of the plurality of data units and an error tolerance range of the first feature stream. By means of the solution, the terminal performs feedback on the basis of the transmission condition of the plurality of data units and the error tolerance range of the first feature stream, and does not necessarily perform transmission error feedback for a data unit having a transmission error (for example, does not necessarily feed back a NACK for a data unit having a transmission error), thereby reducing the number of retransmissions, and improving communication efficiency.

The invention relates to the technical field of numerical control machine tools, and provides a nonlinear error control method and system for a numerical control machine tool, and the method comprises the steps: constructing a nonlinear error dynamic calculation model fusing the axial speed and acceleration constraints, and traversing the time nodes of an interpolation period between two adjacent cutter location points, obtaining a nonlinear error value of each time node according to the nonlinear error dynamic calculation model, and obtaining a maximum nonlinear error and a corresponding actual position; taking two adjacent cutter location points and the actual position of the maximum nonlinear error as fitting feature points, and constructing a nonlinear error equivalent curvilinear equation by adopting a piecewise quadratic function fitting method; performing error control according to the nonlinear error equivalent curve equation and the error allowable value, if the nonlinear error of the interpolation section meets the precision requirement, not performing optimization, and if the nonlinear error of the interpolation section does not meet the precision requirement, performing interpolation point optimization processing to obtain a theoretical trajectory. According to the invention, the error control accuracy can be improved.

Embodiments of the present disclosure provide a method of forming N-type and P-type source / drain features using one patterned mask and one self-aligned mask to increase windows of error tolerance and provide flexibilities for source / drain features of various shapes and / or volumes. In some embodiments, after forming a first type of source / drain features, a self-aligned mask layer is formed over the first type of source / drain features without using photolithography process, thus, avoid damaging the first type of source / drain features in the patterning process.

The application relates to the technical field of structural design and optimization, and particularly discloses a structural error tolerance model construction method for compartment connection positioning, which comprises the following steps: acquiring real point cloud data of a bolt connection pair through three-dimensional laser scanning, realizing parameterized digital geometric modeling in combination with a deep neural network, and giving material properties and adaptive mesh division; generating a dynamic load sequence based on a vehicle driving condition, applying the dynamic load sequence to the model, and extracting full-field displacement vector data by using a digital image correlation technology; further identifying a micro-slip area of a contact interface, and calculating an energy dissipation index of the micro-slip area; establishing a quantitative mapping relationship between the energy dissipation and fatigue life through experiment calibration, taking a target life as a constraint, iteratively optimizing key design parameters by using a pattern search method, and finally outputting a connection structure scheme meeting the durability requirement.

The utility model discloses a vermicelli machine driving device convenient to disassemble and assemble, which comprises a mounting plate and a servo driver, the front surface of the mounting plate is provided with a mounting groove, the servo driver is arranged in the mounting groove, the two sides of the servo driver are respectively and fixedly connected with a pair of positioning plates, and the positioning plates are arranged in the mounting groove. A positioning groove corresponding to the positioning plate is formed in the inner wall of the mounting groove, the positioning plate is clamped in the positioning groove, a supporting pipe is fixedly connected into the positioning groove, a through hole corresponding to the supporting pipe is formed in the positioning plate, a threaded rod is inserted into the supporting pipe in a threaded mode, and a rotating block is fixedly connected to the top end of the threaded rod in a sleeving mode; the two sides of the through hole are both provided with fault-tolerant grooves corresponding to the rotating blocks, the bottoms of the two ends of each rotating block are both fixedly connected with abutting blocks, the whole structure is simple, field assembly is convenient and fast, and the driver can be conveniently disassembled and assembled during follow-up maintenance operation.

Image super-resolution (SR) refers to the process of recovering high-resolution (HR) images from low-resolution (LR) inputs. Blind image SR is a more challenging task which involves unknown blurring kernels and characterizes the degradation process from HR to LR. In the present disclosure, embodiments of a variational autoencoder (VAE) are leveraged to train a kernel autoencoder for more accurate degradation representation and more efficient kernel estimation. In one or more embodiments, a kernel-agnostic loss is used to learn more robust kernel features in the latent space from LR inputs without using ground-truth kernel references. In addition, attention-based adaptive pooling is introduced to improve kernel estimation accuracy, and spatially non-uniform kernel features are passed into SR restoration resulting in additional kernel estimation error tolerance. Extensive experiments on synthetic and real-world images show that embodiments of the presented model outperform state-of-the-art methods significantly with the peak signal-to-noise ratio (PSNR) raised considerably.